Papers
Wed 26 Aug 2026
441 papers
- ModularSQL: A Runtime Guardrail for the Multiplicity Blind Spot in Text-to-SQLLLM safety and security → Safety alignment, refusal and guardrails
- Visual Representation and History Modeling for Navigation World ModelsWorld models → World models for navigation
- HiPACE: Hierarchical Phase-Boundary Analysis and Controlled Evaluation of Feature Absorption in Sparse AutoencodersMechanistic interpretability → SAE features and their applications
- Certified Predictive Value-of-Advice Gating for Cost-Aware Language-Model Guidance in Reinforcement LearningDeep reinforcement learning methods → RL training of language and agent models
- Delay-of-Gratification as a Multi-Agent Survival Micro-benchmark for Long-Horizon LLMs: Social Exposure, Personas, and Tool Use BudgetsLLM agents and tool use → Social simulation and multi-agent debate
- Evaluation of Multi-Turn Consistency in LLM Agents: Survival Analysis and Failure-Rationale TaxonomyLLM agents and tool use → General LLM agent frameworks and benchmarks
- Evaluating Explanation-Driven Vision-Language Reasoning via Generation Order InterventionsLLM reasoning → Multimodal reasoning +1
- BLADE: Distilled LLM Regularization for Calibrated Knowledge Graph CompletionKnowledge distillation → Knowledge distillation for LLMs
- Delegated Misalignment: How Multi-Agent Structures Amplify LLM Safety RisksLLM agents and tool use → Agent threats, guardrails and governance +2
- What Does Chain-of-Thought Entropy Measure? A Channel Audit of Scaffolding, Routing, and ContentLLM reasoning → Prompting and chain-of-thought reasoning
- Multiple latent orderings better predict language model preferencesLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
- Beyond Raw Context Transfer: Representation-based Federated Retrieval-Augmented GenerationMultimodal LLMs / vision-language models → Document understanding and multimodal RAG +1
- StepKV: Step-Aware KV Cache Compression for LLM AgentsLLM agents and tool use → Software engineering coding agents +1
- PAGE: Partition-Aware Gated KV-Cache EvictionEfficient LLM inference and serving → KV cache optimization
- The Limits of Speculation: Bounding Speculative Decoding in Mixture-of-ExpertsMixture-of-experts → Efficient MoE training and inference systems +1
- Beyond the Stitching Assumption: A Unified Framework for Multimodal Synthetic Data Evaluation via Semantic QuantizationSynthetic data and self-improvement for LLMs → Privacy-preserving tabular data synthesis
- Multilingual Safety Signals Are Multi-Layered: Filtering Safety-Degrading Data for Safer LLMsLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
- Towards Scaling Reinforcement Learning to Massive Populations: Learning Mean-Field RepresentationsDeep reinforcement learning methods → RL for games and combinatorial optimization
- Grounded Checklist Partial Credit for Agent Skill TrajectoriesLLM agents and tool use → Agent skill learning and evolution
- Active Curriculum Refinement for Reinforcement LearningDeep reinforcement learning methods → Deep RL algorithms and representations
- Don't Overthink, Don't Underthink: Toward Adaptive Reasoning in Agentic AILLM reasoning → Reasoning for agents, planning and code +1
- SpeechGym: An Audio-Native Gym for Training Voice Agents via Reinforcement LearningLLM agents and tool use → General LLM agent frameworks and benchmarks +1
- The Latent Diagnostic Taxonomy: A Framework for Constructing Classifiers and Diagnosing Their Decisions, Applied to Prompt Injection DetectionLLM safety and security → Prompt injection and agent security
- Interpreting Latent Protein Language Model Features with Geometric AnnotationsMechanistic interpretability → Interpreting biomedical and science models
- Co-Evolving Structured Knowledge and Reasoning in Language ModelsLLM reasoning → Knowledge- and tool-augmented reasoning +1
- Why RAGs Hallucinate: Penalty-Aware Evaluation of Retrieval-Augmented Generation Systems with Knowledge-Gap CanariesRetrieval-augmented generation → RAG robustness and adaptive retrieval
- Assessing the Downstream Utility of Evidence-Aware Retrieval in RAGRetrieval-augmented generation → RAG robustness and adaptive retrieval
- Survival-Guided Length Control for Efficient Diffusion Language ModelsDiffusion models → Discrete and text diffusion
- Knowledge-Verified Emergent Deception in LLM Agents Under Conflicting IncentivesLLM agents and tool use → Social simulation and multi-agent debate
- Neuro-symbolic PRM: Enhancing Scientific Reasoning via Structured Traces and Symbolic VerificationLLM reasoning → Reasoning benchmarks and evaluation
- Privacy Without Regret: Differentially Private Inference-Time AlignmentLLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment
- Approved Too Late: Verdict Staleness in LLM-Guarded Self-Adaptive SystemsLLM safety and security → Prompt injection and agent security
- MemToC: Benchmarking Memory-Tool Conflict Resolution in Large Language ModelsLLM agents and tool use → Agent memory systems
- SKILL.state: Scalable Long-Horizon Agent SkillsLLM agents and tool use → General LLM agent frameworks and benchmarks
- WALL-SS: Scaling Long-horizon World Models via Next-Scale AutoregressionWorld models → Embodied and robot world models
- Procedura: Agentic 3D Modeling with Procedural ControlLLM agents and tool use → Multimodal and embodied agents
- How Do LLM Agents Actually Get the Flag? Trace-Level Provenance for Agentic Offensive Security EvaluationLLM agents and tool use → Agent threats, guardrails and governance
- The Reasoning Tax: Token Economics of LLM Reasoning Across Task Types and Deployment ContextsLLM reasoning → Test-time compute and efficient thinking
- LLM Agents for Time-Series: A SurveyLLM agents and tool use → Financial trading and forecasting agents
- NeuronFuzz: Safety Neuron Guided Fuzzing for LLM Safety EvaluationLLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks +1
- Prompt Sensitivity of Generative Agents: Evidence from an Epidemic ModelLLM agents and tool use → Social simulation and multi-agent debate
- Same Model, Different Harness: Different Coding-Agent ResultsLLM agents and tool use → Software engineering coding agents
- Surgical Video Generation From Diffusion to World Models: A SurveyWorld models → Medical and surgical world models
- VBVR-Pro: A Scalable and Verifiable Suite for Native Visual ReasoningLLM reasoning → Multimodal reasoning
- Zero-WAM: In-Context World-Action Modeling from Human Videos for Open-Ended Task GeneralizationWorld models → Embodied and robot world models
- Agentic Autoresearch for Cell-Edge Power Control: Radically Redefining the Researcher's RoleLLM agents and tool use → Agentic AI for wireless networks
- Finding and using interpretable latents in a neutrino foundation model with sparse autoencodersMechanistic interpretability → Sparse autoencoder training and evaluation
- TraceML: An Empirical Analysis of Human-Agent Planning in Machine Learning DevelopmentLLM agents and tool use → Web, search, and research agents
- SwarmWorld: Stigmergic technological evolution in societies of language-model agentsLLM agents and tool use → General LLM agent frameworks and benchmarks
- Prefix Sliding for efficient test-time scalingLLM reasoning → Test-time compute and efficient thinking
- StreamPI: Streaming Multimodal Temporal Modeling for Vision-Language-Action ModelsVision-language-action models → Efficient VLA inference and adaptation
- One Policy, Many Embodiments: Unified Camera-Centric Action Geometry Pre-training for Heterogeneous Embodied ManipulationVision-language-action models → Human-video and latent-action pretraining
- R3: Training Robots to Reason in Natural Language via Reinforcement LearningLLM reasoning → Spatial and embodied reasoning +1
- Robust CurveMoE: Multi-Norm Adversarial Defense for Mixture-of-Experts Models via Mode ConnectivityMixture-of-experts → Vision mixture-of-experts
- Trace Integrity for LLM Data Agents: A Vision for Auditable Structured Reasoning in Real-World SystemsLLM reasoning → Test-time compute and efficient thinking +1
- DualOPSD: Adaptive Privileged Teachers for On-Policy Self-DistillationKnowledge distillation → On-policy and self-distillation for LLMs
- VISA: Agentic Self-Evolving Data Synthesis for Multimodal Instruction FollowingLLM agents and tool use → Agent skill learning and evolution +1
- Imitation Learning for Connection-Tableau ConstructionOffline RL and imitation learning from data → Imitation learning methods
- A Self-Evolving Multi-Agent Framework Defense against LLM Jailbreak AttacksLLM agents and tool use → Agent threats, guardrails and governance +2
- AsymSpec: Context-Asymmetric Speculative Decoding for Agentic LLMsLLM agents and tool use → Software engineering coding agents +1
- Uncertainty-Guided Latent Diffusion Models for Faithful Super ResolutionDiffusion models → Inverse problems and image restoration
- ProgRouter: Online Progress-Guided Orchestration for Multi-Agent LLM Workflows under Quality-Cost TradeoffsLLM reasoning → Reasoning for agents, planning and code +1
- Multi-Granularity Context-Enhanced RAG over Multimodal Knowledge GraphsMultimodal LLMs / vision-language models → Document understanding and multimodal RAG +1
- SciMIF: Understanding Multimodal Instruction Following in Scientific DomainsMultimodal LLMs / vision-language models → MLLM training and visual instruction tuning
- LivingRAG: Augmenting Graph RAG with ExperienceRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
- 4DGS-WAM: Bridging Past and Future with an Object-Centric World Action Model based on 4D Gaussian SplattingWorld models → Embodied and robot world models
- When Pruning Meets Interpretability: Preserving Sparse Autoencoder Robustness in LLMsMechanistic interpretability → SAE features and their applications
- Candidate supply and answer selection shape the value of LLM judging in multi-agent systemsLLM agents and tool use → Social simulation and multi-agent debate
- One Symptom, Three Levers: A Critical Review of On-Policy Self-DistillationKnowledge distillation → On-policy and self-distillation for LLMs +1
- TAU-Agent: An Agentic Retrieval-Augmented Framework for Traffic Anomaly UnderstandingRetrieval-augmented generation → Multimodal RAG +1
- Code World Model: Coding Agent as World BrainWorld models → Video-generation world models and simulators +1
- Repair or Resample? Rethinking Failure Debugging in LLM Multi-Agent SystemsLLM reasoning → Reasoning for agents, planning and code +1
- Beyond the Editing Canvas: Evidence Divergence in OOXML-to-LLM IngestionRetrieval-augmented generation → RAG robustness and adaptive retrieval
- Do Vision-Language Models Agree on the Affective Qualities of Shape? A Cross-Model Audit for Generative Design InterfacesMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
- VISTA: Visually Inferred Spatial ConTact Attention for Contact-Rich ManipulationOffline RL and imitation learning from data → Robot manipulation imitation learning
- MA-VLA: Multi-Arm Vision-Language-Action Model for Collaboration and Compositional GeneralizationVision-language-action models → Generalist VLA models and benchmarks
- Learning Late, Guiding Early: Timestep-Decoupled Semantic Guidance for Fair Face GenerationDiffusion models → Image generation, text-to-image and editing
- Precipitation Downscaling Using Foundation Model-Conditioned DiffusionDiffusion models → Flow matching, time series and science
- DEFUSE: Generalizable Backdoor Defense for Self-Supervised Encoders with Generative PriorsDiffusion models → Robustness, privacy and adversarial aspects +1
- THA-Flow Generative Model: Prosthesis Geometry Prediction from Preoperative CTDiffusion models → Medical imaging and segmentation
- SkillShield: Prompt-Space Security Skills for LLM Coding AgentsLLM agents and tool use → Agent threats, guardrails and governance
- LocalLSTC: A Long Short-Term Control Architecture for Locally Deployed GUI AgentsLLM agents and tool use → GUI and mobile agents
- EVOMAL: Self-Poisoning in Self-Evolving Coding AgentsLLM agents and tool use → Agent threats, guardrails and governance +1
- ToST: A Tree-of-Thought Socratic Teaching Framework for Multi-Path Guidance and Parallel ThinkingLLM reasoning → Prompting and chain-of-thought reasoning
- Drift-Aware Multimodal User Representation Learning via Multi-Scale Temporal Modeling and Sparse Mixture-of-ExpertsMixture-of-experts → MoE for recommender systems
- HypoForge: A Self-Improving Multi-Agent Framework for Automated Hypothesis Generation and Testing via Scientific Skill LearningLLM agents and tool use → Web, search, and research agents
- Beam Search, Self-Consistency, and the Limits of Inference-Time Scaling for Grammar-Constrained Text-to-SQL in Small Language ModelsLLM reasoning → Knowledge- and tool-augmented reasoning
- TailSFT: Filtered Fine-Tuning Improves Post-Training PerformanceLLM agents and tool use → Reinforcement learning for LLM agents
- Pointing the Way, Hiding the Destination: Practical Private Dense Retrieval at ScaleRetrieval-augmented generation → Privacy and security of RAG
- When RAG Fails to Equalize: Geo-bias in Factual Question Answering over Public CompaniesRetrieval-augmented generation → RAG robustness and adaptive retrieval
- Reassembling Distributed Risk: Trajectory-Conditioned Action Generation for Multi-Turn Agent SafetyLLM agents and tool use → Agent threats, guardrails and governance
- Fairness-Aware Test-Time Prompt TuningMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
- Unsupervised Anatomical Feature Learning via Diffusion Models: Enhanced Medical Image Segmentation with Denoising Diffusion Probabilistic ModelsDiffusion models → Medical imaging and segmentation
- CloSeR: Unified Relational Distillation from Closed-Set Teachers for Category DiscoveryKnowledge distillation → General knowledge distillation methods
- AI Slop and Hallucinations in Vulnerability Assessment: A Survey on Reasoning Failures and Trustworthy MitigationLLM reasoning → Reasoning for agents, planning and code
- Overview of SHROOM-Visions 2026: A Shared Task on Hallucination Detection in Large Vision-Language ModelsMultimodal LLMs / vision-language models → Hallucination mitigation in LVLMs
- From General Agents to RCA Experts: A Self-Evolving Harness for Root Cause AnalysisLLM agents and tool use → Software engineering coding agents +1
- Think-Probe-Respond: Improving Large Language Models as Judges of Research Idea NoveltyLLM reasoning → Knowledge- and tool-augmented reasoning
- GaussianDream++: Efficient 3D Gaussian World Modeling for Robotic ManipulationWorld models → Embodied and robot world models +1
- Towards Purified Multi-Label Test-Time Adaptation of Vision-Language ModelsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
All 441
- Diffusion Transformers for Roof Graph Synthesis and ReconstructionDiffusion models → 3D, point cloud and scene generation
- MAMA-FLUX.2: Image-to-Image Synthesis of Post-Contrast Breast DCE-MRI for the MAMA-SYNTH ChallengeDiffusion models → Medical imaging and segmentation
- A Token-Level Analysis of Sampled-Token Reverse-KL On-Policy DistillationKnowledge distillation → On-policy and self-distillation for LLMs
- AutoVerifier: Residual-Guided Non-Parametric Optimization for Reference-Based Answer VerificationLLM reasoning → Reinforcement learning for reasoning
- AWM: Answerable Working Memory for Long-Document VQA AgentsLLM reasoning → Multimodal reasoning +1
- Advantage-Driven Explicit Memory for Social NavigationOffline RL and imitation learning from data → Social robot navigation
- From Specialization to Generalization: Instruction-tuned LLMs for Robust Harmful Content MitigationLLM safety and security → Safety alignment, refusal and guardrails
- Knowledge Distillation for Efficient Acoustic Echo ControlKnowledge distillation → Speech and audio distillation
- JIT-Agent: Scaling Harness Intelligence via Just-in-Time Harness EvolutionLLM agents and tool use → Software engineering coding agents
- RA-VLA: Retrieval-Augmented VLA for Test-Time AdaptationVision-language-action models → Reinforcement learning and post-training +1
- GRIP: Granular Reward-Guided Parameter Interpolation for Efficient ReasoningLLM reasoning → Test-time compute and efficient thinking
- V-Rubrics: Visual Faithfulness via Rubric-Based Reinforcement LearningMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
- MLLMCLIP: Feature-Level Distillation of MLLM for Robust Vision-Language RepresentationsMultimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP
- ConfAL-WM: Confidence-Guided Active Learning for Action-Conditioned World ModelsWorld models → Embodied and robot world models
- Beyond Scaling: Self-Evolving LLM Agents for Hardware Kernel Optimization via an Experience-Driven Workflow and Experience Graph MemoryLLM agents and tool use → Software engineering coding agents +1
- When Stale Constraints Go Unchecked: Budgeted Verification Failures in Inherited Agent MemoryLLM agents and tool use → Agent memory systems
- Reflection Steering: Disentangling Reflection from Reasoning in Activation Space for Token-Efficient InferenceLLM reasoning → Test-time compute and efficient thinking
- ClueWeaver: Reward-Guided Dual-Agent Evidence Reasoning for Compact LLMs on Literary Long NarrativesLLM reasoning → Reasoning benchmarks and evaluation +1
- Video-IFBench: Evaluating Instruction Following of Multimodal LLMs in Video Understanding ScenariosMultimodal LLMs / vision-language models → Video understanding with MLLMs
- TOPAS: Workflow-Aware Prefix-State Scheduling for Multi-Agent LLM ServingLLM agents and tool use → Software engineering coding agents +1
- Agentic Game Development as a Verifiable Trajectory Data Engine for Scaling World ModelsWorld models → Video-generation world models and simulators +1
- Conditional Total Correlation and the Serial Depth of Adaptive Parallel SamplingDiffusion models → Discrete and text diffusion +1
- CaSKG: Counterfactual-Causal Skill Graphs for Scalable Agent Skill RetrievalLLM agents and tool use → Agent skill learning and evolution
- MMJailBench: A Factorized Benchmark for Disentangling Multimodal Jailbreak VulnerabilitiesLLM safety and security → Vision-language and text-to-image safety
- A Storage-Retrieval Gap in Parametric Knowledge Graph MemoryRetrieval-augmented generation → Knowledge-graph and structured-knowledge RAG
- ReliableRAG: Combating Misinformation in Retrieval-Augmented Generation via Reliability-Guided Reasoning ChainsLLM reasoning → Knowledge- and tool-augmented reasoning +1
- PonsRAG: A Pons-Inspired RAG Bridging Cognitive Islands for Coordinated Long Narrative ReasoningRetrieval-augmented generation → Agentic RAG, agents and memory
- Semi-Supervised Adaptation of Vision-Language Models for Image ClassificationMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
- Homo-RAG: Homology-Guided Retrieval-Augmented Generation for Cross-Species Gene Function PredictionRetrieval-augmented generation → Medical and biomedical RAG
- TailorCoPilot: Enabling Agentic Pattern Making with Version-Controlled State TrackingLLM agents and tool use → Multimodal and embodied agents
- MACGen: Toward Functionally Correct and Secure Code Generation via Multi-Agent CollaborationLLM agents and tool use → Agent threats, guardrails and governance
- MathAdv: What Theorem Provers Know, Reason, Formalize, and GeneralizeLLM reasoning → Mathematical reasoning
- DocPC: Document-Level Visual Retrieval via Representative Page CompositionMultimodal LLMs / vision-language models → Document understanding and multimodal RAG
- Distance Is Not Enough: Forget-Retain Alignment Gap Predicts LLM Relearning RobustnessLLM safety and security → Machine unlearning
- DCGC: Draft-Conditioned Global Correction for Complex Reasoning with Masked Diffusion ModelsLLM reasoning → Diffusion reasoning and test-time scaling +1
- BVR Sim: An Open and High-Throughput Environment for Heterogeneous Air-Combat Reinforcement LearningDeep reinforcement learning methods → RL for robot navigation and UAV control
- Paint What You See: Benchmarking Dexterous Visual Tool Use in Multimodal AgentsLLM agents and tool use → Multimodal and embodied agents
- Can your AI agent be cheaper? Investigating the effects of task specifications on token spend in agentic coding tasksLLM agents and tool use → Agent memory systems
- OmniPhys: A Unified Multimodal Benchmark for Physics Understanding and Generation from Chinese Educational CorporaMultimodal LLMs / vision-language models → VLM benchmarks and visual reasoning
- Refusal geometry reflects refusal training: diverse refusal prefixes can raise stable rank and weaken refusal vector ablation attacksLLM safety and security → Safety alignment, refusal and guardrails
- Adaptive Triggering for Bias Correction in LLM ReasoningLLM reasoning → Reasoning benchmarks and evaluation
- GGSS: Geodesic-Gated Spherical Steering for Inference-Time Debiasing of Generative Vision-Language ModelsMultimodal LLMs / vision-language models → Social bias and fairness in VLMs
- Capacity Overflow: A Blind Spot for Backdoor Attacks in Vision MoEMixture-of-experts → Vision mixture-of-experts
- Where to Look Matters: On-Policy Self-Distillation for Long-Video UnderstandingMultimodal LLMs / vision-language models → Video understanding with MLLMs +1
- Escaping Low-Dimensional Overlap: Multi-Task Model Merging via High-Dimensional Sparse DisentanglementMechanistic interpretability → Sparse autoencoder training and evaluation
- Beyond Pairwise Feedback: Listwise Vision-Language Supervision for Preference-Based Reward LearningMultimodal LLMs / vision-language models → Embodied and 3D vision-language models +1
- HRGuard: Gating Relationship Manipulation in Multi-Turn Agentic AI ConversationsLLM agents and tool use → Agent threats, guardrails and governance
- Not All Attention Heads Contribute to Critical Visual Token Selection: Head-Aware Pruning Matters MoreMultimodal LLMs / vision-language models → Visual token pruning and efficiency +1
- Prefix-Denoising Consistency: Test-Time Verification for Diffusion Language ModelsDiffusion models → Discrete and text diffusion
- V-Link: Recovering Lost Visual Representations in Action DiT for Vision-Language-Action ModelsVision-language-action models → 3D and spatial-aware VLA
- PointRL: Learning Point-Level Vision-Language Grounding from Verifiable Annotation EvidenceMultimodal LLMs / vision-language models → Embodied and 3D vision-language models
- BixBench3: Benchmarking AI agents on research-study-scale computational biology tasksLLM agents and tool use → Medical and scientific domain agents
- Routed Graph Handoff: Adaptive Format Selection for Multi-Agent LLM DelegationLLM agents and tool use → General LLM agent frameworks and benchmarks
- Groundhog Bit-Flip Attack: Seeding Infinite Generation Loops in Mixture-of-Experts LLMs through Bit FlipsMixture-of-experts → Expert routing strategies
- Hierarchical MoE for Multi-Modal ILD DiagnosisMixture-of-experts → Medical and biomedical MoE
- What Do Medical Vision-Language Models Learn in Radiology? Transfer, Alignment, and Source-Proxy Leakage Under Distribution ShiftMultimodal LLMs / vision-language models → Medical vision-language models
- A Few Pages of Markdown: Committed AI Configuration and Lower Quality Cost after Coding-Agent AdoptionLLM agents and tool use → Software engineering coding agents
- LACRM: LLM-Guided Mining of Directed Advertiser Competition Relationships in E-Commerce Advertising
- Seasonal and Quantum-inspired Models for Neutron Monitor Time Series Forecasting
- [DeepGOF] Where Does a Logistic Risk Model Fail? An Audited Neural Goodness-of-Fit Test for Model Development and External Validation
- Task-Resolved Fisher Spectroscopy for Quantum Reservoir Computing
- Classifier-Dependent Benefits of Pseudo-Labeling for Semi-Supervised Android Malware Attribution
- Is Reasoning Always Useful? Rethinking Reasoning Utility in Universal Multimodal Embeddings
- Benchmarking Arabic--Russian Machine Translation: A Comparison of Fine-tuned NMT and Few-shot LLMs under Rich Morphology and Low Lexical Overlap
- UNWIND: Any-Length Facial Video for Stress Detection without Temporal Windowing
- Spaceborne differential photogrammetry for control-free measurement of large-gradient deformation with structural immunity and a predictable accuracy envelope
- StepCOPS: Closed-Testing Lower-Tail Certificates for Language-Model Policy Selection
- Quantifying Protocol-Induced Uncertainty in Comparative Predictive-Model Evaluation: Evidence from Large-Scale Daily PM10 Forecasting
- The Architect, the Adversary, and the Judge: Closed-Loop Generation of Standards-Aligned Assessment Items at Scale
- Quantifying Hidden Salt for Precision Healthcare: Sodium Assessment via Joint-Factor Retrieval and Chain-of-Thought Inference
- Monocultural Biases: Correlated biases in large language models lead to unequal systemic exclusion rates in hiring
- Role-Aware Morgan Fingerprints for Reaction Yield Prediction
- Clustering-Based Collective Anomaly Detection in IoT Systems: A Graph Neural Network Approach
- TARGet: Topology-Aware Fusion-based Radio Frequency Circuit Functional Modeling using Graph Neural Networks
- A Multi-Agent Pipeline for Source-Grounded Synthetic Note Generation from Longitudinal Structured EHR
- MechaTerp-TRACE: A Novel Approach for Component Ablation Analysis in Language Models
- Didactic knowledge or Clinical Cases? How Data Types Shape Medical Large Language Models
- Uncertainty and Business-Aware Remaining Useful Life Estimation for Semiconductor Manufacturing
- From Latent Biomarkers to Clinical Rules: Embedding-Guided Rule Mining and Attribution-Based Translation for Interpretable Tabular Learning
- A Comparative Framework for Evaluating Foundation Models on Tabular Data: A Case Study in Healthcare
- SafeTune: A Unified Faithful Library for Auditing and Repairing Safety Drift in Fine-Tuned LLMs
- Is Imagination Derived from Hallucination? A Cross-Taxonomy Evaluation of Imagination and Hallucination in Large Language Models
- Do Language Models Know Their Own Constraints?
- ECP-Bench: Benchmarking and Learning Entertainment Content Promotion with Foundation Models
- GRRR: The Geometry of Reshaping, Rotation, and Routing in Decoder LLM post-training
- Weak Ties, Strong Signals: Efficient Training Data Detection in Diffusion LLMs via Independent Token Sampling
- Impact of Multiple Non-Invasive Biosignals on Cardiovascular Biomarker Estimation via Simulation-Based Inference
- ES-AHD: An Evolution Strategy Framework for Automatic Heuristic Design
- Defending Wearable VLMs Against Private Attribute Inference
- Projection-Aware End-to-End Learned Video Compression for 360-Degree Video
- FLM: Frequency-Aware Language Models for Generative Image Compression
- Data Diversity, Not Frequency Invariance: A Controlled and Self-Audited Study of Compression-Robust Deepfake Detection
- Distributed Semantic Segmentation With Improved Rate-Distortion Trade-Off
- CARD: Calibration via Agreement in Reverse Diffusion for Out-of-Domain MRI Segmentation
- Multi-Sensor Mapping of Vulnerable Urban Settlements Using SAR, Multispectral, and Hyperspectral Imagery: A Case Study in C'ordoba, Argentina
- Evaluating Constrained Iterative Refinement for Scalable Vector Graphics Generation with Off-the-Shelf VLMs
- Can Tainted Pixels Expose Deepfake Videos?
- Mapping Woody Vegetation from Multi-Source Imagery and Prediction Fusion for Enhanced Data Efficiency and Accuracy
- Compositional Generalization via Structural Identification in a Category-Theoretic Framework
- Diff Mining: Logit Differences Reveal Finetuning Objectives
- Distributed Training using an Intelligent Network
- Toward Equitable Low-Carbon Mobility: Fairness-Aware Demand Prediction for Expanding Bike-Sharing Systems
- Vowel Signs Are Not Letters: A Pre-tokenization Ceiling on Multilingual Tokenizer Fertility
- Subgraph Filtering for Fair Graph Neural Networks
- NeoTriFuse: Reliability-Aware Multimodal Fusion under Missingness Heterogeneity for Neonatal Mortality Risk Prediction
- AfriSwitch: A Benchmark for In-the-Wild African Code-Switched Speech Recognition
- FedCMAPSS: A Benchmark for Federated Learning in Remaining Useful Life Estimation
- Fine-Tuning of Transformer models with Frames
- Redwood: A Frontier AI Accelerator Designed, Verified, and Deployed from Scratch in 2 Weeks by AI
- Towards a universal meta-optics solver via large language models
- Case2Flow: Bridging Patient Cases and Guideline Flowcharts through Multimodal Retrieval
- Simultaneous Envy and Equitability Guarantees
- LowRankArena: A Standardized Evaluation Platform for SVD-Based LLM Compression
- VIPER: An Expert-Curated Benchmark for Vision-Language Models in Veterinary Pathology
- CG4AI: A Column Generation Framework for Training AI Models Under Constraints
- A Unified Framework for the Mechanics of Information in Convolutional Neural Network Image Space
- Cross-lingual Representation Learning via Centroid Intervention Fusion
- Finding the Right Evidence: Factor-Guided Coarse-to-Fine Reasoning for Long Videos
- Cross-simulator transfer with foundation model summaries: Towards robust SKA-era reionization inference
- Decay-Region Group Delay as a Forensic Cue for AI-Generated Impulsive Sounds
- MoganColBERT-TR: A Late-Interaction Multi-Vector Retrieval Model for Turkish
- StreamAV-Bench: A Comprehensive Benchmark for Streaming Audio-Video Generation
- ProofEvolve: Neuro-Symbolic Evolution for Formal Automated Theorem Proving
- Beyond Capability Benchmarks: Learning Operational Fingerprints of LLM Cloud Services from Production Incident Metadata
- How Unlikely Is "Unlikely"? Assessing Verbal Probability Perception Across Large Language Models
- When Is Noise Response Universal? Tokenization as the Hidden Variable in Language Models
- Modality Maturity Index: A benchmark for assessing multimodal capabilities of omni models
- FaithSieve: Fine-Grained Evaluation of Math Proofs with Faithful Formal Evidence
- Algebraic Multigrid Acceleration for Efficient Label Spreading
- District-Level Food Environment Indicators and Social Vulnerability in S~ao Paulo
- On Scope Classification and Current Knowledge-Editing Benchmarks: A Negative Result, with INLAY as a Gradient-Free Case Study
- Assessing mentalization in humans and large language models
- Muon with Finite Newton-Schulz: The Smoothing Benefit in Nonsmooth Nonconvex Optimization
- Multi-Dataset Inverse Problem Solving with Distributed Generative AI
- Constraint-Aware Physics-Informed Neural Networks for Static Shape Estimation of Co-Manipulative Continuum Robots
- 6.5% of the Neuro-Symbolic Literature Can Be Reproduced from Its Published Artifacts, a Six-Stage Audit Framework and First Instantiation
- A causal graph-informed temporal convolution architecture for interpretable retail electricity price forecasting
- Pruning Binarized Neural Networks: A Dedicated Framework and Globally Weighted Algorithms
- Agent Mesh: Reliability Primitives for Non-Idempotent Agent Delegation - Identity Adequacy and Evidence Adequacy
- Classical and Hybrid Quantum Machine Learning for Trigger-Like Event Selection on CMS Open Data: An Eight-Qubit, PCA-Constrained Benchmark
- TRACE: Retrospective Streaming Generation of Physical Fields under Sparse Structured Sensing
- Real-time virtual circuits for plasma shape control via neural network emulators: integration and testing in the MAST-U PCS
- Attention-Guided Reliability Scaling for Contrastive Decoding in Robust Audio-Visual Speech Recognition
- Fairness Invariants: A Relational Approach to Explaining and Mitigating Fairness Bugs
- RefVideo-6M: A Reliable Reference-Based Dataset for Instructional Video Editing
- A Visual Dependence-Aware Framework for Multimodal Unsupervised Continual Post-Training
- MyoMechanix: Biomechanically-Grounded Compositional Skilled Activity Understanding and Coaching
- PlanSightRAG: A Visual-First Multimodal RAG for Automating Question Answering and Compliance Checking for Civil Standard Plans
- Planetary Prediction Engine: Autonomous Geospatial Prediction via Intelligent Data Selection and Foundation Model Embeddings
- ICON decomposition: Multivariate concept auditing reveals shortcuts that neural networks rely on
- Gating Before Commitment: Anticipating Intent Divergence to Prevent Post-Interaction Decision Failures in Autonomous Driving
- Group-Shared Low-Rank Approximation for Mobile-Efficient Pointwise Convolutions in Large-Kernel CNNs
- Fine-Tuning Whisper for Automatic Speech Recognition in Baniwa: A Preliminary Study
- How Much Rank Does LoRA Need? Rank-Error Bounds for Transformer Attention
- The Value of Human Expertise
- Beyond Local Surprise: Grounded Dialogue as Selective Belief Revision under Referential Uncertainty
- UltraPIPS: Improving model perception in B-mode ultrasound with foundation models
- VoiceMem: Streaming Dual-Brain Memory for Real-Time Interaction
- CardioFusion-AI: Robust ECG--PPG Fusion for Multimodal Physiological Monitoring Under Signal Degradation
- Distinct dynamics of conceptual and referential disruptions in human reading and large language model processing
- Spectral Allocation: Why Muon Outperforms Adam, and How to Improve Muon
- FRAME: Separating sampling variation from performance disparities in medical image analysis
- When Personality Meets Quantization: A Layer-wise MBTI Analysis of Quantized LLMs
- Lost but not erased: Finding traces of a forgotten language in neural speech models
- PANDA - Prototype-Anchored Alignment for Partially Unpaired Multimodal Learning, with Applications to Alzheimers MRI and TCGA Pathology
- Quantitative Analysis of ω-Regular Robust MDPs
- Less Contouring, More Accuracy: Lesion-Guided ROI Deep Learning for Ovarian Ultrasound Classification
- Auditable CT Phenotyping Through Report-derived Radiological Observations
- Unveiling Spectral Mechanisms in Training-Free LLM Text Detection
- A Statistical Audit of Physical AI Benchmark Redundancy
- How Robust Are Automated Fact-Checking Systems? A Cross-Benchmark Evaluation
- When Composition Doesn't Add Up: Humans Identifying Defects in AI-Generated Images
- Continually learning neural-operator surrogate for three-dimensional airborne electromagnetic Bayesian inversion
- Controlling for Omitted Variable Bias in Deep Neural Networks
- Formal, Executable and Explainable Runtime Monitoring of Spoken Air Traffic Control Operational Procedures
- Visual General Intelligence: A White Paper
- Query-Side Attacks on GNN-Based KGQA: Tracing Failures from Entity Linking to Answer Generation
- Choose Your Game Wisely: Measuring Game-Theoretic Structures in Real-World Vehicle Interactions
- SAMpLE: A SystemC-AMS Machine LEarning-based Framework for Virtual Prototyping
- Quantum-Inspired Modeling of Driving Behavior
- One Form to Transfer Them All: Pretraining Multilingual Language Models Beyond Native Orthography
- MetaSieve: Faster Relational Deep Learning through SQL-Based Metapath Selection
- Forecasting Multiple Observables with SCROLL: Score-Trained Uncertainty for Stochastic Dynamics
- Towards A Unified Information Bottleneck Framework for Time Series Explanations
- From Passive Response to Proactive Correction: Enhancing LLM Robustness Against Input Fact Perturbations
- A General-Purpose Molecular Foundation Model Transfers Across Diverse Olfactory Tasks
- Scalable Multi-GPU Simulation of 3D Multicellular Growth with RNN-Based Workload Balancing
- Embedding NDRE Trajectories into Contrastive Learning for Label-Free, Physiology-Aware Crop-Stress Staging and DSS Outputs
- Efficient Estimation of High Information Projections using Nearest Neighbours
- Loss-Based Active Learning for Neural Abstractive Summarization
- How Edge of Stability Hinders SCAFFOLD in Federated Optimization
- CEDAR: Controlled and Event-Driven Demand Forecasting via Residual Decomposition
- Anchoring Bias in LLM-as-a-Judge Systems: Prior Scores Compromise Evaluation Independence
- LUTSeg: A Longitudinal Multi-Expert Dataset for Ulcer Tissue Segmentation
- Unlocking Multimodal Protein Language Models at Inference Time
- Key Point Analysis Needs Structure Recovery: Task Definition, Dataset Diagnosis, and a Structure-Aware Benchmark
- Why ML-based cough models do not generalize: a systematic cross-dataset evaluation for tuberculosis screening
- VINCENT: Validated Interaction Network for Cross-drug Explanation of Therapeutics
- Socialized Detector Learning: Trajectory-Guided and Reciprocal Distillation for Heterogeneous Object Detectors
- Missing the Butterfly and Predicting the Past: Features or Bugs of Accurate AI Weather Models?
- Skill Issue: Are Skills Language-Invariant in LLMs?
- FlowMoDL: Model-Based Deep Learning with Conjugate-Gradient Data Consistency for Highly Accelerated 4D Flow MRI Reconstruction
- Unfolding Scientific Papers into Multi-Turn Generation Trajectories for Continued Pre-Training
- Localize-Then-Decide Guarantees for LLM Judgments
- Learning Continuous Regional Temperature Fields with Lead-Time and Resolution Queries
- Steer the Sampling, Not the Kernel Grid: Geometry-Guided Sampling Operator for Volumetric Segmentation
- Mapping the Emergence of Regularization-Driven Dynamics in Grokking
- Label-Free Foundational Model Selection for Medical Image Classification under Distribution Shift via Pseudo Label Discrepancy
- TDFNet: Tri-projection Deformable Fusion Network for Panoramic Salient Object Detection
- Geometry-Constrained Kolmogorov-Arnold Networks: Learning Edge Geometry via Banach Duality
- TacForcing: Streaming Action Generation with Execution-Time Tactile Feedback
- Cooperative Multi-Agent Reinforcement Learning for Adaptive Aggregation in Semi-Supervised Federated Learning with non-IID Data
- EXAONE Tabular 1.0 : Technical Report
- Large Language Model Few-Shot Prompting with Dilemma Training Outperforms Human Surrogates in Predicting Patient Preferences
- MoganBert-TR: A Turkish Encoder Foundation Model Trained from Scratch with a CLM-to-MLM Curriculum
- Learning from waste: Machine Learning for health risk prediction and computer vision-based sorting in Ghana
- LM-X: Explainable Vision--Language--Action Modeling via Progress, Event, and Uncertainty Prediction
- Comparing Corrupted Constrained Learning Problems
- A Constitutive Markov Physics-Informed Neural Operator (MPNO) for Autoregressive Stability in Transient Dynamics
- Why Does Graph Learning Fail to Fully Benefit from a Text Teacher?
- MeMark: Membrane-Space Watermarking for Spiking Neural Networks
- Moving Beyond More Views: Redundancy-Aware Ego-Exo Fusion for Proficiency Estimation
- InteractGesture: Progressive Chunk Guidance for Continuous Streaming Co-Speech Gesture Control
- MIMONet: Multi-scale Input and Multi-scale Output Network for Salient Object Detection
- LongVU-TTT: Causal Test-Time Training for Visual Resampling in Long Video Understanding
- Auditing Privacy Risks in LLM-Enhanced Graph Neural Networks
- It's a matter of timescale: non-linear utility in successor features and multi-objective planning and learning
- Difficulty-Aware Sample Allocation for Adaptive Data Augmentation in Semantic Segmentation
- Multi-output Gaussian process prediction of physical fields under linear equality constraints
- Fast rates in Bayesian online learning with approximate posteriors
- Skeleton-based Zero-Shot Spatio-Temporal Action Localization via Weakly-Supervised Pretraining
- Tropospheric temperature and humidity profile retrieval from Meteosat Flexible Combined Imager based on deep learning
- Modeling spatio-temporal locality in multi-step forecasting of geo-referenced time series
- Adversarial Training of Linear Models under Stealthy Attacks
- Learning New Facts with QLoRA: An Acquisition-Retention Frontier
- Deep Learning Segmentation of Diffusion-Weighted MRI Acute Ischaemic Stroke: A Pragmatic Evaluation Across Three Datasets
- Data Citation for Large Language Models: A Challenge
- Narcissus: Program Synthesis Using Context-Aware LLM Approximations
- Reconstructing the Right Episode: Evaluating Interleaved Conversational Memory Beyond Long Context
- Unmatched Does Not Mean False: Incomplete Reference Sets Can Reverse Calibration Rankings in Open-Ended Theory-of-Mind Tracking
- LDAC-Net: A Learnable Multi-Lag Differencing Attention-Convolution Network for Drift-Robust Recognition with Low-Cost MOX Gas Sensors
- Leveraging Inter-object Affordances for Efficient Planning in Contact-rich Tasks
- DCEO: Direct Causal Effect Optimization for Long-Term User Value Modeling in E-commerce Search
- Propensity Straight-Through Gradients for Discrete Stochastic Systems
- SeVeR: Selective Visual Exposure and Retrieval for 3D Medical Image Question Answering
- Using profiles of cognitive capability to assess AI suitability for workplace tasks
- Plans You Can Check: Verifier-Grounded Learning of an Open-Weight Planner for Executable Video-Editing
- Dissonance Spectrum explicitly models perceptual frequency interactions for better music understanding
- On the Separation of Human and AI-Generated Images in CLIP Embedding Space
- When Should a Network Emit Geometry, and When Should It Detect It? Readout, Reconciliation, and Representation in Floorplan Vectorization
- Nonequilibrium Dynamics of Simple Exclusion Processes Across Dimensions
- Frequency-aware forecasting for short-term typhoon gust prediction
- A Dual-Transformer for Multi-Camera View Recommendation
- M-Fibration Theory with Applications to Weighted Graphs
- A Hierarchical Synergistic Deep Learning Framework Integrating Composition, Structure, and Ionic Transport for Solid-State Electrolyte Discovery
- Individual Fairness in Hierarchical Clustering
- Are Concept Bottleneck Models Effective as Decision-Support Systems?
- Cross-Dataset Stability of Expert-Informed Skill Prompting and Fine-Tuning for Chinese Metaphor Identification
- PolyMemDB: A Polyglot Database System for AI Memory Management
- Generative vs. Encoder Large Language Models for ASR Evaluation: A Comparative Study
- Controllable Affective Generation via Latent Vector Steering
- CrossMambaTuning: Synergistic Spatial and Cross-Layer Adaptation for Machine Vision Compression
- Physics-Informed Foresight Pruning for Sparse PINN Solvers of Nonlinear PDEs
- EgoArgus: Benchmarking VLMs as Situational Assistants for Modality-Grounded User Supports
- AdaVDR: Adaptive Tool Use and Reflection for Video Deep Research
- Virgil: Navigating Explainability for Transformer-based Language Models
- Beyond Optimal Rates in Stochastic Optimization: Trajectory-Adaptive Stopping Rules
- A Hybrid Usability Approach for Rating Evaluation of M-Commerce Applications
- Interpreting Protein Language Model Embeddings via Orthogonal Projection for Protein Fitness Prediction
- A Tendon-Driven Five-Fingered Hand with Distributed Tactile Perception for Dexterous Manipulation
- Goodput Maximization for Large Language Model Edge Inference: A Two-Phase Maskable PPO Approach
- CropCop: An Auditable 120-Class Plant-Health Model from Benchmark Reconstruction to a Quantised Runtime Artifact
- Syn2Logic: End-to-End Neuromorphic Design Automation
- Resilient Decentralized Wireless Federated Learning via Gradient Tracking with AdamW
- Adaptive Hybrid Subspace Levenberg Marquardt Algorithm with Adequacy Monitor for Large Scale Least Squares Problems
- Asymmetric Cross-Modal Fine-Grained Visual Categorization: ACF-Net and the BirdPro Benchmark
- OpenVeinNet: Robust Open-Set Finger Vein Verification with Dynamic Snake Convolution and Graph Learning
- Adaptive Regularization for Random Features: A Neighboring Early-Stopping Rule with Oracle-Rate Guarantees
- A Multi-View Coupled Tensor Decomposition for Lightweight Online Adaptive Traffic Prediction
- FedQoS: Federated QoS-Risk Learning for Heterogeneous Indoor-Outdoor Access Selection
- Pose-Anchored Optical Flow for Low-Latency Human Action Anticipation in Human-Robot Teaming
- SMART: MLLM-guided Temporal Alignment for Unifying Sign Language Recognition and Spotting
- Gaussian Splatting Underwater: A Controlled Cross-Regime Study
- DeCO: Discriminative Evidence Composition for Fine-Grained Dataset Distillation
- 4DStreamCtrl: Interactive Video Generation with Online 4D Control
- VietAIDetector: An Open-Source Zero-Shot Detector for Vietnamese AI-Generated Text
- AERIS: Offline Policy Improvement for Multi-UAV Integrated Sensing and Communication
- PAGS: Autofocusing Photoacoustic Tomography via Speed-of-Sound-Adaptive Gaussian Splatting
- Functional linear regression from sparse to dense designs: a pooling-ridge method and minimax optimality
- Resolving Multi-Modal Regression by Difference-Quotient-Based Clustering:Fast Coarse Conditional-Label Assignment
- Automatic weld seam segmentation for industrial quality control: a comparison of RGB and polarimetric imaging with CNN and transformer architectures
- GLOSS: Geometric Local Self-Similarity Learning for Faithful Reference-Guided Texture Fill
- Training Alignment Auditors via Reinforcement Learning
- VGA-BenchV2: An Expanded Unified Benchmark and Multi-Model Framework for Evaluating Video Aesthetics and Generation Quality
- Energy Yield and Lifetime Climate Classification via Machine Learning for Optimizing Photovoltaic Module Design and Materials
- Joint Initialization of Flux Networks and Effective Multiplication Factor for Physics-Informed Neural Networks Solving Neutron Diffusion Problems
- Saliency-Depth Conditioning for Zero-Shot Segmentation of Communication-Tower Components in Cluttered UAV Imagery
- RotDroid: Cross-Orientation State Equivalence Testing for Detecting GUI Rotation Bugs in Android Apps
- Data-driven Effective Modeling of Stochastic Chemical Reaction Networks
- Bootstrapping a 4D LiDAR Annotation Tool from Video Foundation Models
- AdaptiveEmbed: Sample-Adaptive Multi-Vector Representation for Multimodal Retrieval
- Token-Oriented Semantic Communication with Pretrained Vision Transformers
- PIVOT: A Multi-Trajectory Dataset and Testbed for Pose, Intrinsics, and Novel Viewpoint Evaluation in Real-World 3D Reconstruction
- A meta-algorithm for ab initio reconstruction of complex mixtures in cryo-EM
- Efficient Training with Foresight: Multi-Token Auxiliary Supervision for Autoregressive Image Generation
- Q&A or Document-Based? The Effects of Interface Type on How Screen Reader Users Access Interconnected Documents
- CRAMER: Control via Request-Aware Masking for Editing Recommenders
- RSFusionDet: Underwater RGB-Sonar Multimodal Object Detection
- PaSta: Noisy Node Classification with Partial Label Learning
- FlashNormal: Detailed Surface Normal Estimation from Flash and No-Flash Images
- Leveraging Speech Acts for Low-Data and Cross-Domain Conversation Derailment Forecasting
- Right Values, Wrong Places: Diagnosing and Fixing Record Misplacement in LLM-Generated JSON
- Short Horizons and Sparse Concepts: a Mathematical View of the Readout in the J-lens
- CoRE: Weakly Supervised Coarse-to-Fine Risk Evidence Learning in Driving Videos
- GUIDE: Generative Unsupervised Chinese Query Correction via Phonetic and Visual Shared-ID Encoding
- Provenance Before Prose: Claim-Locked Reporting for Statistical Text Generation
- GraftSR: Grafting Authentic Textures for Real-World Image Super-Resolution via Identical-Instance Guidance
- Learning What to Share and What to Personalize: Hierarchical Strategy Co-Evolution for Agent Memory
- Neither Precision Nor Architecture Alone: Controlled Tests of Failure Remedies for Physics-Informed Neural Networks
- Two Dimensions Govern Agnostic Multiclass Transductive Learning
- FinRiskAtlas: Decision-Aligned Evaluation of Large Language Models for Financial Risk Review
- LLMscope: Extracting LLM Assets from Edge AI Chips via Optical Probing
- Activation-Space Order-Swap Geometry: A Site-Asymmetry Audit
- MulVec: Fine-Grained Role-Aware Matching for Training-Free Zero-Shot Composed Image Retrieval
- SAUSS: Stochastic Approximation with Unbiased Simulated Scores for Limited Dependent Variable Models
- WAVE: Reversing the Guidance Hierarchy for Coarse-to-Fine Guided Depth Super-Resolution
- InsightSR: Refining Symbolic Regression Search Spaces via Parallel Semantic and Structural LLM Guidance
- Combining Self-Embedding Audio Watermarking with Ultra-Low-Bitrate Neural Codecs
- A Training-Free Proactive Defense Against Partial Speech Manipulation via Self-Embedding Steganography
- Generative Action-Chunk Sampling for Adaptive Stiffness Control in Physical Human-Robot Collaboration
- SHSP: Structure-Aware Hierarchical Solution Prediction for Mixed-Integer Linear Programming
- Schr"odinger Bridges over Kinetic Swarming Models
- Provable Non-Acceleration of Standard Strang Splittings of Kinetic Langevin Dynamics
- PhaseShift: Topology-Aware Data Harmonization and Model Consolidation Across Signalized Intersections
- OpenCVL: An Open, Diverse, and Large-Scale Dataset for Fine-Grained Cross-View Localization
- A General Framework for Metropolis-Adjusted Dikin Walks: Dimension-Square Mixing on Polytopes and Log-Det Walks on Spectrahedra
- Mitigating LLM sycophancy with RL-based fine-tuning: Bayesian Truth Serum approach
- Neural-Bayesian Structure Learning for Discrete Choice Modeling
- ShuttleArena: Interpretable Self-Play in Physics-Based Badminton
- The "Curse of Knowledge" in LLM Query Simulation: Concept Provenance for Tracing Answer-Side Intrusion
- From Memorization to Absorption: Mixed-Policy RL for Continual Knowledge Injection